CREB-binding protein (CBP, CREBBP) and the closely related E1A-binding protein p300 (EP300) are transcriptional coactivators that interact physically or functionally with about 10 percent of the estimated 2,000 mammalian transcriptional regulatory proteins. Both genes are required for normal development and physiology, and altered-function- and inactivating-mutations occur in cancer, indicating that CBP and p300 can function both as oncoproteins and as tumor suppressors. Several major unresolved questions remain with regard to the biological and transcriptional roles of CBP and p300. First, it is largely unknown what roles the multiple different transcription factor-binding domains of CBP and p300 play in development and tumorigenesis in vivo. Second, many in vitro studies indicate that CBP and p300 are biochemically indistinguishable, but it is uncertain how redundant their functions are in vivo. Third, numerous in vitro transcription studies have shown that CBP and p300 are crucial coactivators, yet little is known regarding their roles in endogenous gene expression. Fourth, it is uncertain if different classes of coactivators function redundantly. These questions form the core of the application's broad, long-term, objectives, aspects of which will be approached by testing the in vivo functions of two specific protein-binding domains common to CBP and p300. Four strains of knock-in mutant mice have been generated. Two have point mutations on the surface of the KIX domain of CBP and p300 that inhibits the binding of the cAMP-responsive factor CREB and the hematopoietic factor c-Myb, and two strains that have a deletion mutation in the CH1 domain of CBP and p300 that abrogates the binding of the hypoxia-responsive factor HIF-1. Three specific aims will be pursued using these mice. Aim 1 is to elucidate the roles of the KIX domain of CBP and p300 in vivo. Aim 2 is to determine the roles of the CH1 domain of CBP and p300 in vivo. Aim 3 is to establish if CBP and p300 are biochemically redundant in vivo.